In the Summary Table menu, numerical variables are
aggregated into statistical tables, and categorical
variables are aggregated into frequency tables and
contingency tables. And it is visualized to easily
understand the distribution of variables.
Create statistical tables of numeric variables and visualize their distributions.
The summary table function is accessed through the menu system of Descriptive Statistics > Summary Table > Statistical Table of Numerical Variables.
All and Custom with the
radio button.Numerical Variables(Choose one or more):
will appear.User choose is specified in
Target variable selection type:.missing, mean,
standard deviation, skeweness,
kurtosis, observation, SEM,
IQR.mean, standard deviation,
skeweness, kurtosis are selected by
default.min, Q1, median,
Q3, max, 1%th, 5%th,
10%th, 20%th, 30%th,
40%th, 60%th, 70%th,
80%th, 90%th, 95%th,
99%th에서 선택합니다.min, Q1, median,
Q3, max are selected by default.Here, outputting a statistical table from the
Statistical Table of Numerical Variables is assuming that
diamonds data is selected in the Dataset: list
box.
As the default setting of Descriptive Statistics > Summary Table
> Statistical Table of Numerical Variables, if you click the
Execute button, the statistical table is
output in the right result area.
These default settings are:
mean, standard deviation,
skeweness, kurtosismin, Q1, median,
Q3, maxStatistical Table of Numeric Variables
Instead of all numeric variables, the user can select a numeric
variable. This example selects the variables caret,
depth, table and price.
The settings for the statistical table are now defined as follows:
caret, depth, table,
pricemean, standard deviation,
skeweness, kurtosismin, Q1, median,
Q3, maxWhen executed under the above conditions, the following statistical table is output.
Statistical table with selected variables
If you check the Calculate by category check box, you
can calculate the statistical table of numeric variables for each level
of the selected categorical variable.
If you check the Calculate by category check box, a
widget called Categorical variables: is displayed. Here you
select the categorical variable cut.
Select a list of categorical variables
The settings for the statistical table are now defined as:
caret, depth, table,
pricemean, standard deviation,
skeweness, kurtosismin, Q1, median,
Q3, maxcutWhen executed under the above conditions, the following statistical table is output.
Statistical table of numeric variables by categorical varables
Since the statistics table consists of several statistics, it is not easy to understand the distribution of the corresponding numeric variable. However, with visualization, it becomes easier to understand the distribution of numerical variables.
If you check the Plot chart check box, a
density plot is output after the statistical table to
understand the distribution of numerical variables.
The following is the result when the Plot chart check
box is checked. Density plots are visualized for each selected numeric
variable along with a statistical table.
Statistical tables and plots
If you check the Plot chart check box and select a
categorical variable, a density plot of the numeric variable is created,
separated by level of the selected categorical variable.
Statistical table with density plots of numeric variables by level of categorical variables
Create frequency tables of categorical variables and visualize their distributions.
The frequency table function is accessed through the menu of Descriptive Statistics > Summary Table > Frequency Table of Categorical Variables.
All and Custom with the
radio button.Categorical Variables(Choose one or more): will
appear.User choose is specified in
Target variable selection type:.As the default setting of Descriptive Statistics > Summary Table
> Frequency Table of Categorical Variables, if you click the
Execute button, the frequency table is
output in the right result area.
The settings for the frequency table are now defined as:
Frequency table for categorical variables
If you check the Plot chart check box, a
bar plot is output after the frequency table to understand
the distribution of categorical variables.
The following is the result when the Plot chart check
box is checked. Bar plots are visualized for each selected categorical
variable along with a frequency table.
Frequency table with visualization
Compute the contingency table of two categorical variables and visualize the distribution.
The contingency table function is accessed through the menu of Descriptive Statistics > Summary Table > Contingency Table of Categorical Variables.
Type of marginal sum: is displayed.Marginal sum,
Row percentages, Column percentages,
Percentages of total.Marginal sum is selected by default.As the default setting of
Descriptive Statistics > Summary Table > Contingency Table of Categorical Variables,
click the Execute button to output contingency
table for two categorical variables in the right result
area.
The settings for the contingency table are now defined as:
cutcolorContingency table for two categorical variables
You can add marginal sums for aggregation of individual rows/columns to a contingency table.
cutcolorMarginal sumThe contingency table executed under the above conditions is output as follows.
Contingency table with marginal sum
If you check the Plot chart check box, you can output a
mosaic plot that can identify the distribution of two
categorical variables.
The following is the result when the Plot chart check
box is checked. A plot is output to determine the distribution of the
two categorical variables output as a contingency table.
Contingency table with visualization